/, 



505 



-25- 



2. St<Xli» TtST - PtODUCTS I AMP T 



Of Chi eOT^'^MVCQileiio «bo p«relclp«c«d La cta« ••cood CttC Ic wm 

 found tlut 86i%till tWiked the aaae regular brand at at th« cIm of 

 the firat test a /iiK prevlouilf- Tha Min reaaon for change (48X) 

 was one of econoay; t^Tnext aosc important reaaon (13Z) was the Influetx 

 of the tar and nicotine "l^e tables" which bad affected .ainly those 

 with a high coocem for healtls-.:-. Thera bad been a noticeable downward 

 shift in the price distribution oT C«LSP<"><1«*"' regular brands, probably 



«9 a result of the price increase; th'fi'.was trr** of all clusters, but 



'^. ^>%. 



was^eatest for Cluster 6. v^^ 



Inlioly 1975 a further tar and nicotine ^'V^gua table" was issued 



(4), which "sSoy^d that certain major brands on th^^^^.K. aarket, notably 



Players No. 6 ani^^J^abassy Filter, had increased in oi?'(^nc 4«llvery 



froa 1.2 Bg to 1.3 n^^Using the previous arbitrary definitions of 



"low" (less than l.O agj^^jiedium" (1.0-1.2 tg) and "hUh" (1-3 «« *»^ 



above) this caused a ^sid«v»>Jile shift froes the "sediua" to the "high" 



category when respondent*' reguW brands were re-classified on th« new 



data. Table 8 shows the percentage "dUtribution of respondents' regular 



-/; 

 brands across the three nicotine and f out' Jar delivery bands at the 



/*! 

 <; 

 tvote of the 2nd test. - ^/^ 



As in the first test, respondents w«ra"ask^4;>{o rate their regular 



brand for a range of attributes. The ratings are shd^ in Table 9 



which classifies the brands into three groups of different nicotine C 



^^ -- 



delivery. J' ^ • 



-^^ 2 



o 



